Efficient Estimation of Parameters of the Negative Binomial Distribution

Abstract

In this article we investigate a class of moment-based estimators, called power method estimators, which can be almost as efficient as maximum likelihood estimators and achieve a lower asymptotic variance than the standard zero term method and method of moments estimators. We investigate different methods of implementing the power method in practice and examine the robustness and efficiency of the power method estimators.

Keywords:

• Efficiency of estimators
• Estimating parameters
• Maximum likelihood method
• Method of moments
• Negative binomial distribution

Mathematics Subject Classification:

• 62F10
• 62F12
• 62F35
• 62-07

Acknowledgment

We would like to thank the referee for the careful reading of our manuscript and valuable comments.